Development of the Interpersonal Processes of Care Survey–Japanese version

Abstract Aims In the past two decades, there has been growing interest in patient‐doctor communication in psychiatry, and several treatment options have been established. This study aimed to develop the Interpersonal Processes of Care Survey–Japanese version (IPC‐J), which measures multidimensional communication and the relationship between doctors and patients in Japanese psychiatry. Method We conducted a cross‐sectional questionnaire survey at one psychiatric hospital and two psychiatric clinics in Japan and investigated the factor validity, convergent validity, internal consistency, and test‐retest reliability of the IPC‐J. Result Overall, 148 eligible patients participated in the study and were included in the analyses. Data from 16 participants were used to examine test‐retest reliability. An exploratory factor analysis using 23 items from the IPC scale was performed to clarify the factor structure in a Japanese psychiatric setting. The final IPC‐J contained 22 items and a two‐factor structural model. High internal consistency (α > .8) and moderate test‐retest reliability (interclass correlation > .65) were observed. Regarding convergent validity, the factor 1 “Doctor's communication‐related attitudes and skills” was significantly correlated with service satisfaction, empowerment, and medication adherence, whereas the factor 2 “Consideration for the patient's to promote own treatment decisions” was correlated with service satisfaction and medication adherence. Conclusion The IPC‐J appears to be a useful tool for assessing patient views on interpersonal communication with doctors in a Japanese psychiatric setting. While the analysis suggested utilizing an IPC‐J with 22 items, the full IPC‐J can be used in cross‐cultural studies.


| INTRODUC TI ON
In the past two decades, there has been growing interest in patient-doctor communication, and multiple psychiatric treatment options have been developed. 1 For example, shared decision making (SDM) has become an essential skill for psychiatrists and refers to the process by which the doctor and patient discuss and decide on treatments based on the patient's preference and the doctor's professional experience and psychiatric knowledge. 2 Communication between doctor and patient is critical for the successful implementation of SDM. 3 Indeed, a previous study reported that good communication improved patient satisfaction, prognosis of mental illness, and quality of life. 4 In addition, patient-doctor communication was associated with a therapeutic relationship or therapeutic alliance that influenced the SDM process and psychiatric treatment results. 5,6 While several studies have addressed SDM, and the patient-doctor relationship not only in Western countries but also in Japan, 6,7 no studies have examined patient-doctor communication in a Japanese psychiatric setting.
One potential reason is a lack of relevant, validated scales. The Interpersonal Processes of Care Survey (IPC) 8

| Design and settings
We conducted a cross-sectional questionnaire survey at one psychiatric hospital and two psychiatric clinics in Japan to test the factor validity, convergent validity, internal consistency, and test-retest reliability of the IPC-J. Participant eligibility criteria were as follows: (a) receiving outpatient service from the psychiatrists who participated in this study, (b) age 20 years or older, and (c) taking scheduled prescription drugs. We excluded patients with a primary diagnosis of dementia or developmental disorder. Overall, 276 eligible patients were recruited and were informed about the study by research team members or research collaborators who were not involved in the patients' medical care. In particular, potential participants were clearly informed that their psychiatrists would not see their questionnaire responses. It was assumed that consent was given if patients completed the questionnaires, which included the IPC-J, other instruments, and questions regarding background characteristics. A subset of the participants completed the IPC-J after a 2-to 4-week interval to investigate test-retest reliability. This study was approved by the Research Ethics Committee at the National Center of Neurology and Psychiatry (no. A2016-044).

| Characteristics
We asked participants about the following background characteristics: gender; age; educational, marital, and living status; and hospitalization and employment experiences in the past 6 months. We collected information about each participant's main diagnosis and the duration of any outpatient services and obtained data on the presence of comorbid disorders (developmental disorder and mental retardation) from each participant's primary doctors.

| Interpersonal Processes of Care Survey (IPC)
The IPC 8 is a multidimensional, patient-reported instrument that as-

| Other measurements
To examine the convergent validity of IPC-J, we used three scales to assess client satisfaction with outpatient services, empowerment, and medication adherence. The Japanese version of Client Satisfaction Questionnaire 8-item (CSQ-8-J), which was originally developed in the United States, was employed to measure client satisfaction. 9 The internal consistency and convergent validity of the CSQ-8-J were confirmed in a past study. 10 We used the Boston University Empowerment Scale (BUES) to assess empowerment. 11 The Japanese version was developed by Hata et al, 12 who confirmed the internal consistency, test-retest reliability, and convergent validity. Finally, the Medication Adherence Scale (MAS) was used to assess medication adherence. 13 The internal consistency, convergent validity, and factorial validity of this scale were previously confirmed. We hypothesized that IPC-J scores would be positively correlated with those of the CSQ-8-J, BUES, and MAS.

| Statistical analysis
We conducted an exploratory factor analysis (EFA) with geomin rotation to verify the IPC-J construct in a Japanese setting. The number of factors was determined through a scree plot and interpretability. To determine which items belonged to each factor, we extracted the items if they were loaded at a level of ≥0.4 on the factor. For estimation in the EFA, the responses for each item were assumed to be ordinal variables, and the robust weighted least squares method was used to treat the highly skewed distribution of the IPC-J items.
We examined the internal consistency and test-retest reliability of the IPC-J using Cronbach's alpha or intraclass correlation coefficients (ICC), respectively. To determine the convergent validity, Spearman's correlation coefficients were calculated to assess whether the IPC-J positively correlated with the CSQ-8-J, BUES, and MAS.
The EFA was performed using Mplus version 8. 14 Other analyses were conducted using Stata version 15.

| Study participants
We obtained consent for participation from 165 patients (response rate: 59.78%), 20 of whom consented to complete the IPC-J twice for examination of test-retest reliability. After 17 participants with missing IPC-J responses were excluded, a total of 148 participants (53.62%) were included in the analyses of factor structures, internal consistency, and convergent validity. In addition, the data from 16 participants were used to assess test-retest reliability. Table 1 shows the participants' background characteristics.
There were 82 men (55.41%), and the overall mean age was 44.66 (SD = 13.18) years. Over half of the participants had never been married, and over 75% of them lived with their families. Twentyone (14.19%) participants had been hospitalized during the past 6 months, and 65 (43.92%) were employed. Approximately, half of the participants were diagnosed with schizophrenia. Regarding comorbid disabilities, eight (5.41%) participants had a developmental disorder and six (4.05%) were diagnosed with mental retardation.
Participants had been cared for by their primary doctor for a median of 29.47 months (range: 0-382.53 months).

| Factor structure
Confirmatory factor analysis failed to replicate the original factor structure. An exploratory factor analysis using 29 items from the IPC scale was performed to clarify the factor structure for a Japanese psychiatric setting. A two-factor structural model was determined from the scree plot. However, the interpretation of the factor structure was difficult.
To clarify the IPC-J factor structure suitable for a Japanese setting, we discussed the following items. First, items #26, #27, #28, and #29 pertain to patients' impressions of office staff, but this study investigates only patients' relationships with their physicians.
In addition, a previous study suggested that office staff members did not significantly influence patients' evaluations of physician services or overall satisfaction in a Japanese hospital. 15  showed two-and three-factor models. The factor loading value of each IPC-J item exceeded 0.4, except for item #23 of both the twoand three-factor models (Table 3). Finally, we adopted 22 items and the two-factor model (factor 1: "Doctor's communication-related attitudes and skills" and factor 2: "Consideration for the patient's to promote own treatment decisions"), excluding items #23 to #29 from the original scale.

| Reliability
We tested the reliability of the IPC-J two-factor model with 22

| Convergent validity
Spearman's correlation coefficients between IPC-J factor.1, IPC-J factor.2, the CSQ-8-J, the Boston University Empowerment Scale, and the Medication Adherence Scale are shown in Table 4

| Limitation of this study
The limitation of this study was that the participants were only users of two facilities in Japan. In order to verify the scale in service users with diverse diagnosis, it is necessary to conduct surveys that include a variety of facilities and service users.

TA B L E 2
Factor loading for IPC-J (23 items) Bold values emphasize the factors to which each item belongs.

| CON CLUS IONS
In this study, factor analysis was used to develop the IPC-J, which exhibited a two-factor structure. Furthermore, the internal consistency and test-retest reliability of the scale were supported by a high Cronbach's alpha and moderate interclass correlation, respectively. Convergent validity was confirmed via significant correlations with other related scales. These findings suggest that the IPC-J is a useful tool to assess patients' views on the interpersonal aspects of communication with doctors in a Japanese psychiatric setting, while the original full-item IPC (29 items) can be used when conducting cross-cultural studies.  and SY contributed to research planning and data collection, and provided advice regarding the analysis. All other authors contributed to data collection and interpretation, and critically reviewed the manuscript. All authors approved the final version of the manuscript and agree to be accountable for all aspects of the work, ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

DATA D E P O S I TO RY
Not all the data are freely accessible because no informed consent was given by the participants for open data sharing, but we can provide the data used in this study to researchers who want to use them, following approval by the ethics committee of the National Center of Neurology and Psychiatry.

I N FO R M E D CO N S E NT
Informed consent was obtained from all participants.